針對網際網路上的視訊資料流的需求日益增加而發展的，MPEG-4加入了視訊資料流方案。他提供可精密可粗糙的彈性編碼架構 (FGS)，且將此架構與時間可伸縮性編碼架構合併可以解決許多在網路上傳遞視訊所必須面臨的問題。在本論文中，我們發展一套模式決策方法。此方法根據能夠很容易地由基層編碼器獲取的資訊來判斷一個視訊單位在FGS最適合的編碼模式。
現階段在網際網路上傳遞視訊資料最主要的瓶頸在於網路的變動性(如，頻寬的時變性、封包遺失、網路壅塞)，針對這個問題MPEG-4在FGS中提出採用位元平面編碼的可調式編碼傳輸架構，因此只需要同一個事先編碼好的視訊資料就可以在各種不同的頻寬下傳輸，如此較一般彈性編碼架構更能有效率地使用網路頻寬。FGS也有許多衍生性的架構，每一種架構均提供不同程度的位元率分配，且位元率分配可以在畫面與畫面之間(例如，FGS與FGST)或畫面裡的物體與物體之間(例如，selective enhancement與 background composition)。首先我們根據一些在基層編碼器中可容易獲得的資訊來描述一筆視訊資料的特性(例如，畫面複雜度、運動程度)。接著根據這些資訊我們將解決傳統分類問題的方法做適當的變形，來對不同類型的視訊資料在各種FGS的衍生架構中選取一種最適合的編碼模式，且壓縮後的結果均經由一「時間--空間的失真量測機制」(Spatial-Temporal Distortion Metric)來評估視訊的品質。在我們所提出的方法中，整個模式決策過程僅需計算出出現機率最大的模式而事前機率(prior probability)與條件機率(conditional probability)於訓練過程(training process)中已知，因此模式決策所需的計算量甚低，適用於及時處裡的應用。根據實驗結果，所提出的模式決策機制能有效地對不同類型的視訊選取出最適合的壓縮模式。為了更精準地選擇合適的模式，將來我們必須研究更接近人眼知覺品質的衡量機制來評估視訊資料在各種壓縮模式下的失真程度。

Streaming Video Profile is the subject of an Amendment of MPEG-4, and is developed in response to the growing need on a video-coding standard for streaming video over the Internet. It provides fine granularity scalability (FGS), and its combination with temporal scalability addresses a variety of challenging problems in delivering video over the Internet. In this thesis, we develop a mode selection method that may find the most suitable scalable coding from six coding schemes: FGS, FGST, FGS-SE and FGST with background composition based on the information that can be easily extracted from base layer encoder.

Reference
[1] “CODING OF MOVING PICTURES AND AUDIO,” ISO/IEC JTC 1/SC 29/WG 11 N4350, July 2001.
[2] W. Li, “Overview of Fine Granularity Scalability in MPEG-4 Video Standard,” IEEE Trans. Circuits Syst. Video Technol., vol. 11, pp. 301-317, March 2001.
[3] H. M. Radha, M. van der Schaar and Y. Chen, “The MPEG-4 Fine-Grained Scalable Video Coding Method for Multimedia Streaming Over IP,” IEEE Trans. Multimedia, vol. 3, pp. 53-68, March 2001.
[4] M. van der Schaar and H. Radha, “A Hybrid Temporal-SNR Fine-Granular Scalability for Internet Video,” IEEE Trans. Circuits Syst. Video Technol., vol. 11, pp. 318-331, March 2001.
[5] M. van der Schaar and Y.-T. Lin, “Content-Based Selective Enhancement for Streaming Video,” in IEEE Int. Conf. Image Processing, vol.2, pp. 977-980, 2001.
[6] H. Katata, N. Ito and H. Kusao, “Temporal-Scalable Coding Based on Image Content,” IEEE Trans. Circuit Syst. Video Technol., vol. 7, pp. 52-59, Feb 1997.
[7] S. Wolf and M. H. Pinson, “Spatial-Temporal Distortion Metrics of in-service Quality Monitoring of any Digital Video System,” in SPIE Int. Symp. Voice, Video, and Data Communications, Boston, MA, Sept. 11-22, 1999.
[8] D. S. Turaga and T. Chen, “Classification Based Mode Decisions for Video over Networks,” IEEE Trans. Multimedia, vol. 3, pp. 41-52, March 2002.
[9] N. Vlassis and A. Likas, “A Greedy EM Algorithm for Gaussian Mixture Learning,” IAS tech. Report, Sep 2000.
[10] T. K. Moon, “The Expectation-Maximization Algorithm,” IEEE Signal Proc. Mag., vol. 13, pp. 47-60, Nov 1996.
[11] W. Li, “Bit-plane coding of DCT coefficients for fine granularity scalability,” ISE/IEC JTC1/SC29/WG11, MPEG98/M3989, Oct. 1998.
[12] B. Schuster, “Fine granular scalability with wavelets coding,” ISE/IEC JTC1/SC29/WG11, MPEG98/M4021, Oct. 1998.
[13] S.-C.S. Cheung and A. Zakhor,, “Matching pursuit coding for Fine granular scalability,” ISE/IEC JTC1/SC29/WG11, MPEG98/M3991, Oct. 1998.
[14] H. M. Radha, Y. Chen, K. Parthasarathy and R. Cohen, “Scalable Internet video using MPEG-4” Signal Processing: Image Comm., pp.95-126, 1999.